Statement of David O. Carson, General Counsel, United States Copyright Office Library of Congress 101 Independence Avenue, S.E.

Washington, D.C. (202) 707-8350 Before the SUBCOMMITTEE ON COURTS, THE INTERNET AND INTELLECTUAL PROPERTY OF THE HOUSE COMMITTEE ON THE JUDICIARY 108th Congress, 2d Session July 15, 2004 Mr. Chairman, Mr. Berman, and distinguished members of the Subcommittee, I appreciate the opportunity to appear before you on behalf of the Copyright Office to testify on internet streaming of radio broadcasts. In my testimony today, I will address the workings of the section 114 compulsory license and the role the Copyright Office has played in administering this license. As you know, in 1995, Congress passed the Digital Performance Right in Sound Recordings Act of 1995 (“DPRA”) 1 which, for the first time, granted to copyright owners of sound recordings an exclusive right to make public performances of their works by means of certain digital audio transmissions, subject to a compulsory license for certain uses of these works codified in section 114 of title 17 of the United States Code. In the Digital Millennium Copyright Act (“DMCA”)2 of 1998, Congress updated section 114 and expanded the scope of the compulsory license. We at the Copyright Office believe the creation of a limited performance right in sound recordings was a step in the right direction. It has fostered the growth of new digital technologies which support the legitimate use of music transmitted in digital networks such as the Internet and satellite radio

1 2

Pub. L. No. 104-39, 109 Stat. 336 (1995). Pub. L. No. 105-304, 112 Stat. 2286 (1998).

services. However, there are those who still oppose a public performance right in sound recordings and would oppose any further expansion of that right beyond the limited performance right granted to the copyright owners by virtue of the passage of the DPRA and the DMCA. Whether to expand the scope of the performance right or limit it further remains the prerogative of Congress. But we are convinced that after considering the current state of affairs and the workings of the section 114 statutory license, Congress should be reassured that the creation of a digital performance right, although limited in its scope, was the proper step to take at that time in order to strike a workable balance between the rights of the copyright owners and the demands of users who wished to use these works in new and creative ways. In fact, technological advances since the DMCA was enacted in 1998 pose new threats to performers and sound recording copyright owners, and this hearing provides an opportune occasion to reconsider the scope of the performance right for sound recordings and whether it offers sufficient economic incentives for the investment in and creation of sound recordings in light of the threats posed by the emergence of additional new technologies that threaten to transform activities such as digital broadcasting into interactive enterprises that may further weaken the traditional market for distribution of sound recordings Background Sound recordings did not receive protection under the 1909 Copyright Act or under earlier versions of the copyright law. Instead, a copyright owner had to seek relief at common law in state courts for unlawful use of their works. That changed in 1971 when Congress enacted a law, effective February 15, 1972, that granted exclusive rights of reproduction and distribution to copyright owners of sound recordings.3 Congress took this action in order to curb the mounting losses suffered by the record

Sound Recording Act of 1971, Pub. L. 92-140, 85 Stat. 391 (1971).


industry from the burgeoning trade in pirated records and tapes. However, Congress did not grant the full bundle of rights given to other copyright owners because traditional users of these works fiercely opposed a performance right for sound recordings. Moreover, the more limited set of rights seemed sufficient to deal with the immediate problem of record piracy. Even so, those who opposed federal copyright protection for sound recordings mounted a constitutional challenge to the amendment adding a limited copyright for sound recordings. Twice, the courts considered the question and in both cases the courts upheld the law as constitutional, 4 confirming the position long held by the Copyright Office that a sound recording was capable of being considered the “writing of an author” within the constitutional sense5 and reinforcing the conclusion that sound recordings are creative works worthy of full copyright protection. 6 Although these events settled the basic question of copyrightability and questions with respect to the reproduction and distribution rights for sound recordings in the early 1970's, the debate on whether and to what extent sound recordings should enjoy full federal copyright protection that began in the 1960's has continued. In most cases, stakeholders have retained their original positions during the intervening period, although there is now a general consensus that performers and record producers’ creative contributions are entitled to some degree of copyright protection.

See Shaab v. Kleindienst, 345 F. Supp. 589 (D.D.C. 1972) (sound recordings qualify as writings of an author that may be copyrighted); Goldstein v. California, 412 U.S. 546 (1973) (the term “writing” can be broadly interpreted by Congress to include sound recordings). See Supplementary Register’s Report on the General Revision of the U.S. Copyright Law, House Comm. Print (1965) at 51 (“1965 Supplementary Register’s Report”) (“We believe that, leaving aside cases where sounds have been fixed by some purely mechanical process involving no originality whatever, the aggregate of sounds embodied in a sound recording is clearly capable of being considered the “writing of an author” in the constitutional sense. ... Thus, as indicated in the 1961 Report, we favor extending statutory copyright protection to sound recordings.”). See Statement of Barbara Ringer, Register of Copyrights, before the Subcommittee on Patents, Trademarks and Copyrights of the Committee on the Judiciary, United States Senate, pursuant to S. Res. 72 on S. 111, July 24, 1975, at 11(“July 1975 Statement of the Register of Copyrights").
6 5



Historically, television and radio broadcasters, jukebox operators, and wired music services–the traditional users of the sound recordings who publicly perform sound recordings–have opposed any changes to the Copyright Act that would require payment of a royalty for the performance of a sound recording. These users were already paying authors and publishers of musical works for the right to perform the musical works embodied in sound recordings and saw no reason to make a second payment to performers and record companies for the same performance. Traditional users, however, did not stand alone in their opposition to the movement for a full performance right. In the early 1960's, music publishers aligned themselves with these users and opposed the public performance right for sound recordings because they feared that the creation of a sound recording public performance right would result in a decrease in their stream of revenue. Basically, they envisioned that the royalty pool generated from the public performance of recorded music would remain fundamentally the same and that they would have to share these royalties with the record companies and the performers of sound recordings. On the other side of the debate stood the representatives of the record companies – e.g., the Recording Industry Association of America (RIAA) – and representatives of the performers – e.g., the American Federation of Musicians (“AFM”). The record company representatives took the position that there was no principled reason for treating sound recordings differently from other categories of works. AFM took a broader view. It focused more sharply on the economic deprivation experienced by performers who received no compensation from the public performance of their own recordings, while others, including jukebox operators, radio and television broadcasters and wired music services as well as composers and music publishers- benefitted commercially from these actions. However, AFM did offer a solution to the problem in 1967, during the early stage of the debate regarding the revision of the 1909 Act. It proposed an amendment to establish a “special performing right that would


endure for 10 years and would be subject to compulsory licensing,”7 a novel idea that would not come to fruition in any form until thirty years later. Copyright owners and performers were not alone in their quest for the elusive performance right. On a number of occasions during consideration of the omnibus bill to revise the 1909 Copyright Act and since, the Copyright Office has voiced its unwavering support for the creation of a full performance right for sound recordings, while also acquiescing to proposals to subject the right to a compulsory license.8 In fact, the push for a performance right nearly paid off. Proponents were successful in getting Senator Harrison Williams to introduce a formal amendment to the 1967 Senate bill which, among other things, aimed to create a compulsory license for the public performance of sound recordings. The amendment was accepted when the revision bill was reported by the Senate Subcommittee on Patents, Trademarks and Copyrights to the full Judiciary Committee on December 10, 1969, and remained in the 1971 and 1973 bills, which were reported favorably by the full Senate Judiciary Committee on July 3, 1974. The amendment, however, did not survive opponents’ efforts to remove the provision from the bill, and it was removed from the 1975 revision bills in both the Senate and the House. In fact, the issue was so explosive that in 1975, Register of Copyrights Barbara Ringer refrained from pushing for the creation of even a limited public performance right for sound recordings in the omnibus bill, and testified accordingly: At the same time it must be said that, on the basis of experience, if this legislation were tied to the fact of the bill for general revision of the copyright law, there is a danger that it could turn into a “killer” provision

See Second Supplementary Report of the Register of Copyrights on the General Revision of the U.S. Copyright Law: 1975 Revision Bill, October-December 1975, pp. 214-216 [Draft] (“Register’s Second Supplementary Report”). See 1965 Supplementary Register’s Report; July 1975 Statement of the Register of Copyrights; and Register of Copyrights, Report on Performance Right in Sound Recordings, H.R. Doc. No. 15 (1978) (“1978 Report on Performance Right in Sound Recordings”).



that would again stall or defeat omnibus legislation. This danger exists even more clearly than when I testified to this same effect last July, and would be very severe if the potential compulsory licensees–notably the broadcasting and jukebox industries–exerted their considerable economic and political power to oppose the revision bill as a whole. Should this happen, there could be no question about priorities. The performance royalty for sound recordings would have to yield to the overwhelming need for omnibus reform of the 1909 law.9 Thus, when Congress passed the 1976 Copyright Act, it did not include a performance right for sound recordings. It did, however, ask the Copyright Office to submit a report on January 8, 1978, making recommendations as to whether Congress should amend the law to provide performers and copyright owners any performance rights in sound recordings. But change could not occur in a hostile environment. In that report, the Copyright Office reaffirmed its earlier position and stated without qualification that a right of public performance for sound recordings is fully warranted, offering the following explanation for its unwavering position: Such rights are entirely consonant with the basic principles of copyright law generally, and with those of the 1976 Copyright Act specifically. Recognition of these rights would eliminate a major gap in this recently enacted general revision legislation by bringing sound recordings into parity with other categories of copyrightable subject matter. A performance right would not only have a salutary effect on the symmetry of the law, but also would assure performing artists of at least some share of the return realized from the commercial exploitation of their recorded performances.10

Testimony of Barbara Ringer, Register of Copyrights (December 4, 1975), before the House of Representatives, Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the Committee on the Judiciary on H.R. 2223, Serial No. 36, part 3, at 1908 (1975).


1978 Report on Performance Right in Sound Recording, at 177.


The predicate underlying this position – that the creation and delivery of music requires a joint effort by songwriters and music publishers as well as performers, record producers and record companies – was not widely recognized in the early 1960's, and even in the early 1970's certain opponents of the performance right continued to argue that sound recordings lacked sufficient creativity to justify copyright protection.11 Nevertheless, the realization that the creation and delivery of music had changed dramatically over time and was the result of the contributions not only of composers and music publishers but also of performers and record producers gradually took hold, becoming a generally accepted principle by 1978, and one which remains unquestioned today. Yet, in spite of this general understanding and the efforts of those who supported a full performance right for sound recordings, no legislation was passed in response to the Office’s 1978 recommendation, and the controversy died down. The debate remained relatively dormant until the late 1980's. Congress acknowledged that the development of digital audio tape (“DAT”) machines posed a real threat to the record industry and passed the Audio Home Recording Act of 1992 (“AHRA”).12 Congress passed AHRA to allay the fears of copyright owners that consumers would use the new technology to make unauthorized high - quality digital reproductions en masse, thus displacing sales in the marketplace.13 It did so by requiring the incorporation of a Serial Copy Management System into each digital audio recording device in order to prevent serial copying, and by requiring payment of a royalty fee for the importation and distribution, or manufacture and distribution, of digital audio recording

11 12 13

See Register’s Second Supplementary Report at 221. Pub. L. No. 102-563, 106 Stat. 4237 (1992). H. Rep. No. 102-873, at 18-19 (1992).


media and devices. AHRA also immunizes a consumer who has made a noncommercial reproduction of a musical recording as provided in Chapter 10 of Title 17 from suit for infringing the reproduction right of the copyright owners, although it does not transform infringing consumer uses into non-infringing ones. And it does not cover reproductions of songs stored on a computer in which one or more computer programs are fixed. But use of DAT recorders was merely the tip of the iceberg. Digital technology continued to advance at a rapid pace, forcing Congress to reexamine the effect of new digital technologies on the record industry. The outcome of this reevaluation was an acknowledgment from Congress in 1995 that the advent of on-demand digital subscription services and interactive services posed a serious threat to performing artists and record companies. Record companies believed, and rightfully so, that consumers would adapt to the new technologies and use these services to fulfill their desire to obtain music, and do so without having to purchase a retail phonorecord. Consequently, after carefully weighing the rights of the copyright owners against its desire to foster new technologies and business models, Congress took action in 1995 and passed the Digital Performance Right in Sound Recordings Act (“DPRA”), which granted copyright owners of sound recordings an exclusive right to perform their works publicly by means of certain digital audio transmissions, subject to certain limitations. In taking this action, Congress sought to preserve and “protect the livelihoods of the recording artists, songwriters, record companies, music publishers and others who depend upon revenues from traditional record sales, ... without hampering the arrival of new


technologies, and without imposing new and unreasonable burdens on radio and television broadcasters, which often promote, and appear to pose no threat to, the distribution of sound recordings.”14 For this reason, the DPRA restricted the application of the new digital performance right to interactive services and subscription services, and specifically exempted traditional over-the-air broadcasts and related transmissions, including certain retransmissions of radio signals and incidental transmissions and retransmissions made to facilitate an exempt transmission. It created these exemptions in recognition of the fact that the possibility of these transmissions displacing sales was never very high. It also included a statutory license for subscription services so that these services could avoid the difficulties involved in direct licensing and devote more of their resources to developing new business models for the benefit of the public. However, services operating under the statutory license are subject to specific terms that are designed to limit unauthorized copying of the works by the recipient of the performance. These terms include requirements that the service avoid the use of a signal that would cause the receiver to change from one program to another; refrain from publishing or preannouncing particular songs that will be played during the course of a program; and schedule songs to avoid playing too many different songs by the same artist or from the same phonorecord in a short period of time or, to state it in legal terms, to avoid violating the “sound recording performance complement.” While these terms did offer a measure of protection to copyright owners and performers during the early days of the technological era, they only covered those problems associated with services in existence at the time. It soon became apparent that the DPRA was too narrow. It failed to anticipate

S. Rep. No. 104-128, at 14-15 (1995).


the rapid development of the Internet and its ability to offer perfect digital transmissions to a global audience instantaneously. Thus, three years later, Congress had to revisit the issue of digital audio transmissions and consider how the digital performance right applied to new non-interactive, nonsubscription services that were springing up overnight and offering real time transmissions of a wide variety of musical choices over the Internet to anyone who had a computer. These services, commonly referred to as webcasters, offered for the first time a rich and diversified selection of music for free over a communications network that was readily accessible to anyone with an internet connection. The problem, however, was the unique programming options that these services offered. For example, some webcasters offered “artist-only” channels that played works of one artist continuously 24 hours a day, while other webcasters offered programming techniques that permit listeners to influence the selection of sound recordings that are part of programs created by the webcasters.”15 In light of these programming capabilities and the exponential growth of these new services, Congress recognized that even nonsubscription services can pose a threat to the economic health of the record industry. For this reason, it again amended section 114 with the passage of the DMCA to clarify that the digital performance right applied to these non-subscription webcasters and that these services came within the scope of the statutory license. Moreover, Congress imposed additional terms, beyond those already adopted under the DPRA, on these new nonsubscription services in order to address the programming and technological problems raised by Internet transmissions.

Committee of the Judiciary House of Representatives, Section-by-Section Analysis of H.R. 2281 as passed by the United States House of Representatives on August 4, 1998, at 52 (Comm. Print 1998).



Specifically, the expanded section 114 license requires licensees: to cooperate with copyright owners to prevent recipients from using software or devices that scan transmissions for particular sound recordings or artists;16 to allow for the transmission of copyright protection measures that are widely used to identify or protect copyrighted works;17 and to disable copying by a recipient in the case where the transmitting entity possesses the technology to do so, as well as taking care not to induce or encourage copying by the recipient.18 Congress also made a few other modifications to the Copyright Act in 1998. One major change was the creation of a second statutory license in section 112(e). This license allows any service operating under the section 114 statutory license to make one or more ephemeral recordings 19 of a sound recording to facilitate the digital transmissions of these works governed by section 114. The DMCA also differentiated between those services that were operating prior to the passage of the 1998 amendments and those that came on line after the DMCA’s date of enactment, October 28, 1998. The three preexisting subscription services (Music Choice; DMX Music, Inc.; and Muzak, L.P.) and the two preexisting satellite digital audio radio services (Sirius Satellite Radio, Inc. and XM Satellite Radio, Inc.) comprise the former group and all other services fall into the latter category. Prior to the DMCA, the rates for the preexisting services were set in accordance with four statutory objectives that also apply to

16 17 18 19

17 U.S.C. 114(d)(2)(C)(v). 17 U.S.C. 114(d)(2)(C)(viii). 17 U.S.C. 114(d)(2)(C)(ii).

These reproductions are referred to as ephemeral copies because they generally must be destroyed within six months of the first transmission to the public.


some of the other statutory licenses but do not necessarily yield a marketplace rate.20 These services retained this standard when section 114 was amended in 1998 even though Congress adopted a willing buyer/willing seller standard for setting rates for all other services operating under section 114. Congress’s responses to threats from new digital technologies in 1995 and in 1998 were limited, just as in 1971. Each time, Congress has chosen to focus only on the immediate problems presented to it and to calibrate the rights of sound recording copyright owners to address these particular problems, rather than adopt a full performance right, even though many urged Congress to grant sound recording copyright owners a full performance right. In testimony before the Senate Judiciary Committee in 1995, the Register of Copyrights restated the Office’s steadfast support for a full performance right for sound recordings, citing the need to harmonize the rights for copyright owners of sound recordings with those of the music publishers once and for all.21 Moreover, an earlier study conducted by the Copyright Office in 1991 had underscored the need for such a right as a means to protect record companies and performers who suddenly were faced with the high probability that digital technology would provide readily available

Section 801(b)(1) provides that “rates applicable under sections 114(f)(1)(B), 115, and 116 shall be calculated to achieve the following objectives: (A) To maximize the availability of creative works to the public; (B) To afford the copyright owner a fair return for his creative work and the copyright user a fair income under existing economic conditions; (C) To reflect the relative roles of the copyright owner and the copyright user in the product made available to the public with respect to relative creative contribution, technological contribution, capital investment, cost, risk, and contribution to the opening of new markets for creative expression and media for their communication; and (D) To minimize any disruptive impact on the structure of the industries involved and on generally prevailing industry practices.” 17 U.S.C. § 801(b)(1). Statement of Marybeth Peters, Register of Copyrights, before the Senate Committee on the Judiciary (March 9, 1995).



distribution channels for the reproduction and performance of their works without a counterbalancing means to compensate the creators of the sound recordings.22 In light of this danger, there was no principled reason to continue to allow one group – music publishers – to receive compensation for the performance of their works while denying another similarly situated group of copyright owners – record companies – the same right to collect royalties for the very same performance, especially in the case where the users’ businesses relied heavily on the use of the creators’ works to turn a profit. This is an observation that has been made repeatedly in support of a full performance right and one articulated by the Working Group on Intellectual Property Rights in its 1995 report on Intellectual Property and the National Information Infrastructure.23 This report characterized the lack of a performance right in sound recordings as “an historical anomaly that does not have a strong policy justification–and certainly not a legal one. Sound recordings are the only copyrighted works that are capable of being performed that are not granted that right.”24 Nevertheless, most users of these works continue to oppose a full performance right for sound recordings and argue that the economies in the current marketplace favor the user and the emerging technologies over the creator, even those who stand on the opposite side of the argument when it is their works that are being targeted for use by another group. Indeed, in the last few weeks, broadcasters have participated in meetings at WIPO considering proposals for a treaty that would obligate countries

Report of the Register of Copyright, Copyright Implications of Digital Audio Transmission Services (October, 1991). Information Infrastructure Task Force, Intellectual Property and the National Information Infrastructure: The Report of the Working Group on Intellectual Property Rights (Sept. 1995).
24 23


Id. at 222.


to provide exclusive rights to broadcasting organizations against the fixation, rebroadcasting and retransmission of their broadcast signals, among other rights. The broadcasters claim this new protection is necessary due to changes in technology, such as the Internet, which threaten their existing business models. They seek these rights notwithstanding their efforts here in the United States to oppose and limit the same rights for the creators of the sound recordings that the broadcasters transmit. Paradoxically, if such a treaty is concluded, broadcasters may be able to exercise exclusive rights over their performance of sound recordings even though the copyright owners of the same sound recordings have no rights in that context. Congress has the power to remedy this situation and strike the proper balance in favor of a full performance right. Thus, the question should no longer be whether Congress should provide a full performance right for sound recordings, but rather whether it should be subject to statutory licensing and, if so, what the value of that right should be in order to insure that copyright owners and performers have sufficient monetary incentives to continue to create works for the enjoyment of the public, and what restrictions, if any, should be placed on that right to insure the viability of new businesses to disseminate the works in a high-quality, readily accessible format. Stated another way, the challenge of copyright in this context, as it is in general, is to strike the "difficult balance between the interests of authors and inventors in the control and exploitation of their writings and discoveries on the one hand, and society's competing interest in the free flow of ideas, information, and commerce on the other hand."25


Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417, 429 (1984).


The Section 114 Statutory License –- How it affects broadcasters Although the digital performance right enacted in 1995 and expanded in 1998 is a step in the right direction, it is not an unfettered right. It is subject to certain exemptions – e.g., nonsubscription broadcast transmissions are exempt – and to a statutory license for certain noninteractive transmissions. Pursuant to this license, many digital transmissions of performances of sound recordings may be made without the permission of the copyright owner if the licensee adheres to the terms of the license, pays the statutory royalties, and complies with the Copyright Office regulations governing notice and recordkeeping. Users, however, have complained that the license terms and regulatory requirements have in some cases created barriers that prohibit them from taking advantage of the license. a. Scope of the exemption for nonsubscription broadcast transmissions .

Broadcasters have been particularly vocal about their treatment under the license, arguing in the first instance that they should not be subject to the digital performance right for their digital, Internetbased activities, such as webcasting. At the outset of the first rate setting proceeding for the webcasting license, broadcasters argued that retransmissions of AM/FM broadcast programming enjoyed an exemption from the newly created digital performance right and that simulcasts of radio broadcast programming therefore were not subject to the statutory license. The recording industry and associations representing the interests of performers26 did not agree. They opposed this interpretation and sought a ruling from the Copyright Office declaring that retransmissions of a broadcast signal over a digital

RIAA represented the interests of the record industry in the rate setting proceeding and the rulemaking proceeding to address the legal questions regarding the scope of the section 114 statutory license as it relates to simulcasts of broadcast radio programming over a digital communications network, like the Internet. The Association of Independent Music, the AFM, and the American Federation of Television and Radio Artists filed comments jointly with the RIAA in the rulemaking proceeding.



communications network, such as the Internet, were not exempt from the digital performance right under section 114(d)(1)(A) of the Copyright Act, as amended by the DMCA. Because the resolution of this question would determine whether broadcasters chose to participate in the rate setting process and because it was necessary to resolve whether the rates being set would apply to broadcasters’ retransmissions over the Internet, the Copyright Office postponed the rate setting hearing until it could decide the legal questions posed by the broadcasters and the record industry. Broadcasters, however, questioned the Office’s authority to conduct a rulemaking to ascertain whether simulcasts of AM/FM broadcast programming over the Internet came within the scope of the section 114 statutory license. For this reason, the National Association of Broadcasters (“NAB”) filed an action in the U.S. District Court for the Southern District of New York, seeking a declaratory ruling on the issue.27 This action was eventually withdrawn. In the meantime, the Copyright Office conducted a notice and comment rulemaking proceeding and made a determination that the exemption for broadcast transmissions did not include transmissions made over a digital communications network such as the Internet.28 The key question in this proceeding centered on the meaning of the phrase, “nonsubscription broadcast transmission,” which is not defined expressly in the law. More specifically, the analysis focused on the statutory definition of the term “broadcast” transmission. The statutory definition characterizes a “broadcast” transmission as “a transmission made by a terrestrial broadcast station

27 28

See NAB v. RIAA, 00-CV-2330 (S.D.N.Y.).

65 Fed. Reg. 77292 (Dec. 11, 2000) (amending the regulatory definition of a “Service” in order to clarify that transmissions of sound recordings by means of digital audio transmissions over a communication network, such as the Internet, are not exempt from copyright liability under section 114(d)(1)(A) of the Copyright Act).


licensed as such by the Federal Communications Commission.”29 The Office then focused on the phrase “licensed as such by the FCC,” finding that it limited the exemption to those transmissions made under a license issued by the FCC, and that these transmissions are limited to the local service area of the radio transmitter. In reaching this conclusion, the Office noted that Congress used the descriptive term “overthe-air” frequently in the legislative history to identify those broadcasts that it sought to protect under the exemption and never referenced any other type of transmission made by an FCC-licensed broadcaster when discussing the scope of the exemption. In addition, the Office determined that had Congress wished to exempt all transmissions made by an FCC-licensed broadcaster – the position urged by the broadcasters – then there would not have been a need to carve out additional exemptions to cover certain retransmissions of an AM/FM radio broadcast program. In reaching this conclusion, the Office focused on an exemption in the law which provides that the performance of a sound recording by means of a digital audio transmission is not an infringement in the case of a retransmission of a radio station’s broadcast transmission, provided that “the radio station’s broadcast transmission is not willfully or repeatedly retransmitted more than a radius of 150 miles from the site of the radio broadcaster.”30 Broadcasters had argued that this150-mile exemption applied only to third parties who retransmitted the original broadcast programming and not to the original broadcaster, but the Office rejected this interpretation. The law draws no distinction between the original broadcaster and third party retransmitters, nor does it or the legislative history offer any reason why Congress would allow
29 30

17 U.S.C. § 114(j)(3). 17 U.S.C. 114(d)(1)(B)(i).


original broadcasters to retransmit their programming globally while at the same time restricting the retransmissions of others to a defined geographic area. In fact, an exception in the law to the 150-mile limitation for retransmissions of a radio signal in the case where the radio signal is “retransmitted on a nonsubscription basis by a terrestrial broadcast station, terrestrial translator, or terrestrial repeater licensed by the Federal Communications Commission”31 supports this position. In all cases, the purpose of these provisions is to restrict each retransmission of a digital audio transmission of a radio signal to a limited geographic area, even in those instances where the retransmissions are done by terrestrial physical facilities regulated by the FCC. The Office found further support for its determination that broadcasters could not retransmit AM/FM radio programming over the Internet when it examined section 112, the provision that governs the making of ephemeral copies of sound recordings necessary to facilitate a public performance under the section 114 statutory license. While traditional broadcasters can make a single server copy of their radio programs to facilitate their over-the-air broadcasts under an exemption in section 112(a), webcasters are unable to rely upon this provision for making all the necessary ephemeral recordings that are needed to facilitate a transmission over the Internet. Webcasting requires more than a single copy of a work to effectively transmit over the Internet. For this reason, Congress created a second statutory license in section 112(e) which, subject to the rates and terms of the statutory license, allows a webcaster operating under the section 114 statutory licensing regime (or certain services that provide transmissions to a business establishment for use during the normal course of business) to make one or more ephemeral recordings to facilitate their transmissions. Thus, broadcasters who wish to retransmit

17 U.S.C. 114(d)(1)(B)(i)(I).


their radio station programs over the Internet would have to operate under the section 114 license in order to be eligible under the section 112(e) statutory license to make all the ephemeral recordings needed to effectuate the retransmission of the AM/FM radio program over the Internet. Not surprisingly, the broadcasters did not accept the Office’s determination. They immediately filed a lawsuit under the Administrative Procedure Act challenging the Register’s determination, but the Register’s decision was upheld by both the district and the appellate courts.32 In making its decision, the United States Court of Appeals for the Third Circuit rejected the broadcaster’s fundamental argument that Congress had intended to provide a broad exemption to cover any transmission made by a licensed broadcaster. Specifically, it held that the reference to “broadcast station” in the definition of a “broadcast” transmission referred to the physical facility licensed by the FCC and not to the broadcaster. It noted that under the FCC rules a station must be a physical facility and that the FCC license referenced in the statutory definition must be tied directly to the operation of a particular facility rather than a corporate entity. Consequently, the court held “[a] ‘broadcast transmission’ under § 114(d)(1)(A) would therefore be a radio transmission by a radio station facility operated subject to an FCC license and would not include a webcast. AM/FM webcasting does not meet the definition of a ‘nonsubscription broadcast transmission’ and does not therefore, qualify under § 114(d)(1)(A) for an exemption from the digital audio transmission performance copyright of § 106(6).”33 The court found additional support for its conclusions in the fact that Congress included additional exemptions from the digital audio transmission performance right for retransmissions of certain
32 33

Bonneville Int’l. Corp. v. Peters, 347 F.3d 485 (3rd Cir. 2003), aff’g 153 F. Supp.2d 763 (E.D. Pa. 2001). Id. at 495.


nonsubscription broadcast transmissions, noting that the common-sense reading of the exemptions in § 114(d)(1)(B) requires an interpretation that does not differentiate between webcasting of AM/FM radio programming by one group, i.e, broadcasters, and webcasts of the exact same programming by third parties. Likewise, the court read the legislative history of the DPRA and the DMCA as supporting an exemption for traditional radio broadcasts, and concluded that the exemption for a “nonsubscription broadcast transmission,” which was added with the passage of the DPRA in 1995, did not contemplate protecting AM/FM webcasts by any group. This interpretation of the scope of the exemption for “nonsubscription broadcast transmissions” offered by the Office and by the courts is totally consistent with Congress’ perception at the time the DPRA was enacted that traditional over-the-air radio did not pose a threat to the record industry. b. Interactive services.

The section 114 statutory license is not available to an interactive service. Such a service is defined, in general, as “one that enables a member of the public to receive a transmission of a program specially created for the recipient, or on request, a transmission of a particular sound recording, whether or not as part of a program, which is selected by or on behalf of the recipient.”34 Interactive services must negotiate separate licenses in the marketplace with the copyright owners of the sound recordings for the right to perform publicly specific sound recordings by means of a digital audio transmission.

The statutory definition provides additional explanatory language to distinguish between interactive and non-interactive services, stating that “[t]he ability of individuals to request that particular sound recordings be performed for reception by the public at large, or in the case of a subscription service, by all subscribers of the service, does not make the service interactive, if the programming on each channel of the service does not substantially consist of sound recordings that are performed within 1 hour of the request or at a time designated by either the transmitting entity or the individual making such request. If an entity offers both interactive and noninteractive services (either concurrently or at different times), the noninteractive component shall not be treated as part of an interactive service.” 17 U.S.C. § 114(j)(7).


Congress took this position and imposed full copyright liability on interactive services because Congress realized these services had the greatest potential for displacing record sales. Consequently, in 2000 the Digital Media Association (DiMA) petitioned the Copyright Office to initiate a rulemaking proceeding for the purpose of adopting an amendment to the rule defining the term “Service” to make it clear that a service is not interactive simply because it offers the consumer some degree of influence over the programming offered by the webcaster. After considering DiMA’s arguments for initiating the rulemaking and RIAA’s opposing arguments, the Office determined that a rulemaking was not the appropriate way to resolve the question of interactivity because there was no way to articulate with any precision specific guidelines that would distinguish between an interactive service and an non-interactive service beyond what was already in the statute, especially when business models were undergoing constant change.35 Moreover, the Office noted that “such a determination had to be made on a case-by-case basis after the development of a full evidentiary record in accordance with the standards and precepts already established in the law.”36 Consequently, the Office denied the petition. c. Notice and recordkeeping requirements.

Sections 114(f)(4)(A) and 112(e)(4) require the Librarian of Congress to establish regulations specifying notice and recordkeeping requirements for use of sound recordings in a digital transmission. Accordingly, the Office issued interim regulations on March 11, 2004, specifying notice and recordkeeping requirements for use of sound recordings under the sections 112 and 114 statutory
35 36

65 Fed. Reg. 77330 (Dec. 11, 2000). Id. at 77332.


licenses.37 These rules require users of the section 112 and 114 statutory licenses to report on the sound recordings they perform so that SoundExchange, the collective that collects the statutory royalties and disburses them to copyright owners and performers, knows how to divide up the royalties for performances of sound recordings. Because the amount of royalties paid to each copyright owner and performer depends upon the number of performances of each sound recording, such reporting is crucial to the operation of the statutory license. Requirements have long been in place for preexisting subscription services, and we believe they are working well.38 However, the rulemaking proceeding governing notice & recordkeeping requirements for eligible nonsubscription services such as webcasters is ongoing, and it has proved to be difficult and controversial. Representatives of record companies and performers have sought comprehensive information about each and every performance of each and every sound recording transmitted by a service, arguing that such information is essential in order to ensure that the correct amount of royalties is paid to each copyright owner and performer, and that information that will permit monitoring compliance with the requirements of the sound recording performance complement is also needed. Webcasters and broadcasters opposed such detailed reporting requirements, asserting that they would be excessive and too onerous for an industry that historically has accounted for its performances of musical works in a totally different manner. Throughout the rulemaking, they maintained that the Office should require reporting of only that information that would identify the sound recording for purposes of making a distribution of royalties. Specifically, they submitted that only five data elements would be needed for
37 38

69 Fed. Reg. 11515 (March 11, 2004). See 63 Fed. Reg. 34296 (June 24 1998).


this purpose : name of the service, sound recording title, name of the artist, call sign of the station and date of transmission. They also suggested that the rules should allow services to obtain this information through a sampling process (e.g., providing information for only two weeks out of every year) rather than accounting for each performance. In adopting interim regulations setting the requirements for the information that eligible nonsubscription services must report to SoundExchange, we rejected the type of sampling proposed by broadcasters because it would be likely to under report – or omit reporting at all – performances of the lesser known artists and performers receiving playtime from those webcasting services that offer multiple channels of niche programming, covering an array of genres, e.g., hip-hop, gospel, classical, country, folk, new age, and pop. Morever, we found it difficult to credit claims from webcasters that although their transmissions – and frequently the programming of the content of their transmissions – are controlled and accomplished by the use of computers, they would be unable to report all actual performances of sound recordings. Ideally, this computer-driven medium should be well-suited to the reporting of actual performance data that would ensure that each copyright owner and performer is compensated for the value of the transmissions of performances of his or her recordings. On the other hand, we recognized that for many webcasters, maintaining and reporting any information at all about their transmission of performances would be a novel experience, and that it would be desirable to have a period of transition during which they would become accustomed to such reporting. Thus, while it is likely that we shall require year-round reporting of all performances in the not-too-distant future, the new interim rules require licensees to maintain records for two weeks out of every quarter, identifying which sound recordings were performed during this period and how often they 23

were performed. In deriving these rules, the Office balanced the need to obtain accurate information about performances of specific sound recordings for purposes of compensating as many copyright owners entitled to receive these fees as possible against the burden imposed on the services to provide the needed information and the need for a period of time during which licensees will become accustomed to reporting actual performance data. The ultimate goal remains a final regulation requiring year-round reporting. Meanwhile, the interim rules require the licensees to report only a relatively minimal amount of specific information needed to identify and differentiate sound recordings from one another. In addition to its own name and the category of transmission (e.g., eligible nonsubscription transmission other than a broadcast simulcast, or eligible nonsubscription transmission of a broadcast simulcast, or eligible transmission by a business establishment service making ephemeral recordings), a licensee is currently required to report as few as four key items for each sound recording performed: sound recording title; featured recording artist, group or orchestra; sound recording identification; 39 and total number of performances.40 They do not require the licensee to report other information sought by the record industry, such as the catalog number, the track label (P) line, the duration of the sound recording, the universal product code, or the release year. Nor are the licensees required to report specific information that would aid the copyright owners in assessing compliance with the programming restrictions, e.g., the

The sound recording identification may consist of either the International Standard Recording Code (ISRC) for the particular recording or, in lieu of the ISRC, the album title and the marketing label of the company that markets the album which contains the sound recording. Total performances may be reported either by reporting the actual number of times a sound recording was performed by the licensee multiplied by the number of recipients; or by reporting the total number of times the sound recording was performed as well as the licensee’s aggregate tuning hours – i.e., the total number of listener hours by all who have accessed the service during a given period of time.


start date and time of the transmission of the sound recording. Moreover, the rules do not require a full census report at this time, although they do require licensees to maintain precise records for two weeks out of every quarter. The rulemaking is ongoing. The Office is still considering rules that would establish specific electronic formats for transmitting this information. The format issue has proven difficult. One might have imagined that although there would be differences of opinion over what kind of information must be reported, the interested parties would be able to work out the technical issues involving the electronic formats in which the reports of use would be made. SoundExchange has been working on its own system for maintaining the data that will be reported to it on sound recording performances, and many broadcasters and webcasters have their own electronic systems that already report information on their performances. We had anticipated that SoundExchange could sit down with broadcasters and webcasters to work out the details of how these systems can communicate with each other, but thus far very little progress has been made despite our encouragement and urging.41 We at the Copyright Office have no familiarity with or expertise about the electronic systems maintained by SoundExchange, broadcasters and webcasters, but the interested parties appear to have decided to leave it to us to prescribe the technical rules on the formatting of reports of use of sound recordings, specifying precise fields and delimiters for reporting the required information. We remain hopeful that the parties may come to an agreement – and we strongly urge them to do so–but meanwhile, we are considering a recent submission from RIAA that proposes revised specifications for filing electronic reports of the

“The Office encourages copyright owners, broadcasters and webcasters to work together to agree on formatting requirements that will serve all of their needs, and to submit joint proposals or comments if possible.” 67 Fed. Reg. at 59576 (Sept. 23, 2002).


performance data and has been forwarded to DiMA for consideration. We hope to publish a notice of proposed rulemaking on formatting requirements this summer, and we are optimistic that we can conclude that phase of the rulemaking proceeding by the end of this year. We are also near to concluding the portion of the proceeding concerning reports of use for the historic period. On Tuesday, we published a Notice of Proposed Rulemaking concerning reporting requirements for use of sound recordings during the period prior to April 1, 2004. The notice proposes use of data already provided by the preexisting subscription services to SoundExchange for the relevant period as a proxy for the reporting of actual performances made by all other services during the same time period. This approach had been suggested in our Notice of Inquiry, 42 and has been endorsed by the copyright owners and performers as well as the affected licensees. Both groups have acknowledged that little useful data exists at this point in time and that there is no apparent way to reconstruct the information needed to file reports of actual use. Consequently, copyright owners, performers and licenses advocate the use of a proxy to account for the historic performances. Use of a proxy, however, is an imperfect solution, since it is likely to undercount some performances and over-count others. Nevertheless, it has many advantages. First, the data from the preexisting services for the historic period offers accurate reporting for programming that is by and large comparable to what was offered by the nonsubscription services during the same time period. Second, the preexisting subscription services had transmitted a diverse number of sound recordings so that a large number of copyright owners and performers can be compensated. And finally, the data has already been used by SoundExchange for distribution of royalties received from the preexisting subscription

68 Fed. Reg. 58054 (Oct. 8, 2003).


services and can easily be used for distribution of the royalties received from the nonsubscription services for the corresponding time period. For these reasons, we believe the use of the reports of the preexisting subscription services as a proxy represents the simplest, most practical and cost-effective solution, and that the affected parties will continue to embrace this solution. Interested parties have thirty days to file comments either in support of this solution or offering alternative proposals. d. Conditions for use of the statutory license.

It is our understanding that, now that the question of whether their Internet transmissions are exempt from the performance right has been resolved against them, broadcasters are questioning whether certain terms in the statutory license should apply to simulcasts of AM/FM programming when retransmitted over the Internet. Specifically, broadcasters have focused on those provisions that prohibit a service from announcing its play schedule in advance and the requirement that a service not play more than a limited number of selections from a particular record album or by a particular recording artist within a 3-hour period (the “sound recording performance complement”). These restrictions, among others, were adopted in 1995 to inhibit copying of music by consumers who could make near-perfect digital copies of a sound recording. The reasons behind the restrictions are simple to understand. They were adopted to make it difficult for an individual to identify in advance, and thereby copy, specific works, thus avoiding the expense of purchasing a copy of the work. The need for such restrictions, however, may be less obvious when one considers a typical radio program offering Top-40 selections. Many radio stations routinely play the same selections over and over so that one need wait only a short time before the most recent release of a hit song is played over 27

the airwaves. Consequently, preannounced schedules of these programs may do little to prevent a listener from copying the newest hits. Thus, it is unclear whether the restriction has much value with respect to these types of radio programs. On the other hand, it is hard to understand how the term creates a hardship for broadcasters who simulcast over the Internet today or to understand the need for such preannounced schedules, since most listeners would not consult a program guide before listening to AM/FM radio anyway. The typical practice is to flip on the radio and surf the channels to see what is playing at the moment or to tune in to a favorite talk show at the regularly scheduled time. Thus, until more information comes to light, it is hard to understand what harm the broadcasters suffer today under the preannouncement restriction, or why there is a need to eliminate this term with respect to broadcast programming. Similarly, it is hard to understand the broadcasters’ complaint with respect to the sound recording performance complement restriction since the definition was crafted so that it would permit programming that was typically used by broadcast radio stations. Specifically, the legislative history notes that “[t]he definition [of the complement] is intended to encompass certain typical programming practices such as those used on broadcast radio.”43 Whatever confusion does exist with respect to the application of this provision may well stem from a misunderstanding of what the complement does and does not allow. For example, it would not prohibit a service from playing the same three songs from a single phonorecord as many times as it wanted during a 3-hour period, provided that no more than two of these songs were played consecutively. The sound recording performance complement would similarly allow a service to play up to four different songs by the same featured recording artist or four

S. Rep. No. 104-128, at 34 (1995).


different songs from any particular boxed set of phonorecords over and over again during a 3-hour period provided that no more than three of these songs were transmitted consecutively. Since these provisions seem to accommodate normal scheduling practices, it is hard to see how the sound recording performance complement imposes a burden on a typical AM/FM broadcast station. Certainly, should these restrictions be shown to pose a substantial burden on programming practices that outweigh whatever protection they provide, then Congress should take another look at their application to broadcast programming being retransmitted over the Internet. In fact, that day may well be near at hand, because new technologies and software that allow a consumer to capture and edit programming transmitted via the Internet already threaten their effectiveness. Digital audio broadcasting – Does it pose a threat to copyright owners? Digital audio broadcasting, also known as HD radio, is no longer a vision of the future. Technology to facilitate digital audio broadcasts has already been approved by the Federal Communications Commission (“FCC”). In 2002, the FCC adopted the in-band on-channel system developed by iBiquity Digital Corporation as the standard technology for enabling digital broadcasts by AM and FM radio stations that wished to begin digital transmissions over the airwaves immediately.44 Although radio stations did not immediately embrace the new technology, they are doing so now. In January of this year, KZIA in Cedar Rapids, Iowa, began the movement when it announced its intent to become the first station to offer HD radio.45 Less than five months later, iBiquity issued another press

Digital Audio Broadcasting Systems and Their Impact on the Terrestrial Radio Broadcast Service, 17 FCC Rcd 19990 (2002). IOWA – First in the Nation for HD Radio, The Hollywood (January 2, 2004) located at



release, announcing that radio station KEMR-FM in San Jose, California, had become the 100th radio station to launch HD radio broadcasts.46 It also has compiled a list of more than 300 licensed radio stations that have begun offering HD radio or will begin to do so soon.47 The electronics industry has also been hard at work. Companies are manufacturing and marketing digital radio receivers for those who wish to be among the first to receive clear, digital radio signals over the airwaves. But technologists have not stopped there. Companies are also busy designing and manufacturing new products to capture and record these signals and anticipate the release of a number of new products which will allow a consumer to record digital audio radio signals so that a listener can listen to his or her favorite radio talk show, news show or music program at a later time. In some instances, these products will operate in the same manner as a VCR or a TiVo device, allowing the listener to fast-forward over the segments that one prefers not to hear.48 In fact, some early digital radio recorders, e.g., Blaze Audio’s Radio Recording Suite,49 already include functions that allow the listener to program the device to record a program at specified times, convert an analog signal into a digital format, and upload the recorded program onto a personal computer in a transferable file. In spite of these features, the early release of these devices did not disturb the copyright community because radio programming was not being offered in a digital format at the source.

HD Radio Going Live Coast-to-Coast ... and Beyond (April 19, 2004) at iBiquity has established a website,, where visitors can find information about stations across the United States that are either offering HD radio now or intend to do so in the near future. See Elisa Batista, A TiVo Player for the Radio, Wired News (May 12, 2003), at http://,1282,58769,00.html.
49 48 47



Consequently, programs that were transmitted in an analog format and later converted to a digital format were only as good as the original analog signal. In many cases, recordings of these signals were plagued by static, fades, and hisses. The advent of digital audio broadcasting (“DAB”) and advances in the recording devices, however, will greatly improve audio quality, removing the flaws associated with analog broadcasts. Moreover, these devices and software packages will allow the listener to change the traditional passive listening experience into an interactive process. They will give the recipient the means to edit and store specific segments and songs from a prerecorded program, upload these selections onto the recipient’s personal computer, and allow for further distribution of these segments to others via electronic transfers over the Internet or by other means. On-Demand Audio expects to offer a digital radio recorder this fall that will provide these functions.50 It promises not only to capture and record the digital radio signal, but also to include technology which will allow the listener to skip from song-to-song and skip over advertisements. Moreover, according to its promotional material, its SongSurfer Technology will be able to identify specific segments of a radio program or a song, and bookmark each segment for identification and use at a later time. The product will also include a Jukebox Mode which will allow the user “to save songs,

See also Neuros HD 20GB MP3 Digital Audio Computer located at: Neuros+Players&product%5Fid=401020.



interesting ads, and talk radio segments to a built-in Jukebox. .... Saved songs can then be sorted into playlists either when they are saved or later.”51 Similar technology is available to capture online music over the Internet. Replay Music promotes its ability to save every song played by an on-line music service, automatically tag each song with the artist name and song title, and separate the song into individual tracks for easy access and play-back. The company claims that its “Replay Music sports the most sophisticated track splitting algorithms on the planet. Besides just recording and tagging, each MP3 file contains the entire song–no more, no less.”52 These technological advances threaten to disrupt the careful balance Congress struck between the record industry, on the one hand, and the purveyors of new digital technologies, on the other, in the DPRA and the DMCA. Moreover, widespread use of these products would alter the longstanding relationship between record companies and radio broadcasters in which record companies have provided radio stations with the latest releases at no cost in exchange for promotional airplay, a relationship based on record companies’ expectation that consumers would purchase new CDs based upon what they heard over the airwaves. But today listeners are not limited to what they hear on the radio to inform their choices, nor do they necessarily purchase CDs containing the songs they like. Instead, new technologies, e.g., peer-to-peer services, offer free access to music and a means to obtain free copies of the works they enjoy. In this new environment, record companies cannot necessarily have any expectation of financial reward because consumers find ways to obtain copies of their works for

On-Demand Radio Overview at http://www.gotuitcom/audio/agradio.html. See also


Replay Music at


free. Nevertheless, radio broadcasters who use music as a hook to get listeners and, by extension, advertising dollars, as well as the makers of the software packages that facilitate the free exchange of music over the Internet profit directly from their use of sound recordings. Clearly, the threat posed by today’s new technologies is most ominous for the performers, the record companies and authorized on-line record stores, like iTunes and MusicMatch, whose profits depend, at least to some extent, directly upon sales of CDs or digital downloads; but the potential harm is not restricted to these businesses. Broadcasters and subscription services will suffer, too, from the use of technologies that can capture, record, and preserve individual sound recordings, and the more valuable segments of a radio station’s program. Subscription services will find it hard to sell reproductions of a sound recording to listeners through use of a “buy button,” when these listeners can capture the songs they want and upload them directly to their personal computers with the use of a OnDemand Audio device or Replay Music software. Why would anyone pay for a reproduction of a sound recording when they can create their own private music collection without expending a dime for the reproduction? Broadcasters could also suffer from extensive use of these new technologies, albeit in a more indirect fashion. In the event that the TiVo type devices become popular, listeners will simply avoid the ads, making it ineffective for businesses to advertise on radio. Were this to occur, businesses will seek better ways to reach consumers, and advertising dollars will no longer flow to the broadcasters. The answer, however, is not to inhibit the roll out of HD radio; nor is anyone suggesting a slowdown on this front. HD radio promises to deliver a high-quality audio product that should draw consumers back to the airwaves. The more promising approach would be to grant copyright owners of the sound recording a full performance right so that they can seek marketplace solutions to the problem, 33

perhaps by negotiating licenses for performance rights that would include measures to protect against the types of activities that would make record sales obsolete. At the moment, sound recording copyright owners have no means to prevent a broadcaster from broadcasting their works over the airwaves or to compel protection of their work. Alternatively, Congress may want to consider technological methods to prohibit unlawful copying, an approach the Federal Communications Commission has already begun to explore. On April 20, 2004, it published a Notice of Inquiry to consider the question of digital audio content control in response to concerns presented to the it by the Recording Industry Association of America. While we take no position on the FCC’s recent action, it is apparent that digital audio broadcasting raises many of the same concerns and fears voiced by the record industry when digital technologies first made their appearance in the nineties, and these concerns are even more valid today. How the issues should be addressed, however, remains an open question. But what is clear is that the process must include a careful analysis of copyright policies. Moreover, any solutions adopted must provide strong incentives to the creators to continue their artistic endeavors and equally strong incentives to encourage the continued development of new technological advances. In the absence of corrective action, the rollout of digital radio and the technological devices that promise to enable consumers to gain free access at will to any and all the music they want will pose an unacceptable risk to the survival of what has been a thriving music industry and to the ability of performers and composers to make a living by creating the works the broadcasters, webcasters and consumer electronic companies are so eager to exploit because such exploitation puts money in their pockets.


Mr. Chairman, as always, we at the Copyright Office stand ready to assist you as the Committee considers how to address the new challenges that are the subject of this hearing.